Fig. 1. Adenoviral vectors can deliver genes to cultured cerebellar granule neurons (CGN). (A) Cell viability assay, CellTiter Glow, showed a linear correlation between measured luminescence and the number of cultured cells. Thus, the assay could be used to determine cell viability following adenoviral infection. (B) At an MOI of 300 and greater, adenoviral infection induced substantial loss of cell viability. However, at 100 pfu/cell and below, adenoviral infection produced no evident toxicity. Thus, MOI of 100 pfu/cell was chosen for the induction experiments. (C) CGN were infected with an adenoviral vector carrying coding sequences of enhanced green fluorescent protein (Ad5-EGFP) at an MOI of 100 pfu/cell, and the cultures were photographed with a confocal microscope 48 h after the infection. A high proportion of the cells were infected with the adenovirus. Uninfected cells served as the negative control for the fluorescence signal and for possible toxicity induced by the viral vector. No toxicity from the viral vector was evident. Thus, adenoviral vectors can infect CGN and express the genes that the viral vectors carry. The experiment was repeated with wild type and nNOS−/− CGN cultures three times with similar results. Magnification bar in C = 100 μm.

Fig. 2. Forskolin activates the cAMP pathway in CGN. CGN cultures were infected with an adenoviral reporter vector containing a luciferase gene coding region under the control of a basic promoter with six tandem copies of the CRE consensus binding sequence (Ad5-CRE-Luc). Twenty-four hours after the infection, the treatment wells received forskolin (10 μM). At 48 h, the cells were harvested, and luciferase activity was determined. Luciferase activity, which reflects cAMP pathway activity, is expressed as relative light units (RLU) per microgram of protein. Data represent the mean of three independent experiments. Error bars represent standard error of the mean. Forskolin led to a 17-fold increase in luciferase activity, indicating that CGN possess a functional cAMP pathway.

Fig. 3. Forskolin, a cAMP pathway inducer, protected CGN cultures against alcohol-induced cell death to a greater extent in wild type cells than in cells deficient for nNOS. Cerebellar granule cultures were established from wild type and nNOS−/− mice. Twenty-four hours after establishment, the cultures were treated with forskolin (0 or 10 μM) for 24 h. The cells were then exposed to ethanol (0 or 400 mg/dl) and forskolin for an additional 24 h, and the percent of alcohol-induced cell death was determined using the trypan blue staining method. Data represent the mean of at least three different litters from each genetic background. Error bars represent standard error of the mean. All treatment groups were significantly different from each other (p < 0.05). In the absence of forskolin, alcohol-induced cell losses were greater in the nNOS−/− mice than in the wild type mice. Forskolin reduced alcohol toxicity in both strains, but more effectively in the wild type cells than in the nNOS−/− cells. Thus, stimulation of the cAMP pathway exerts a neuroprotective effect against alcohol, and this protective effect is linked to the expression of nNOS.

Fig. 4. Forskolin's neurotrophic effect and neuroprotective effect against alcohol toxicity are PKA-dependent. (A) CGN cultures were treated with forskolin in the presence of the selective PKA inhibitor H89 (concentrations ranging from 0 to 10 μM). Forskolin alone exhibited a neurotrophic effect and substantially increased cell viability. H89 at the highest concentration not only abolished this effect, but reduced cell viability drastically. As the concentration of H89 decreased, forskolin's effect was reinstated. Thus, selective inhibition of PKA blocked the neurotrophic effect of forskolin. (B) CGN cultures exposed to alcohol (0 or 400 mg/dl) were treated with forskolin (0 or 10 μM) in the absence or presence of H89 (0, 5 or 10 μM). Alcohol-induced reductions in cell viability were determined. Alcohol alone (in the absence of forskolin or H89) substantially reduced cell viability. Forskolin protected the cells against alcohol toxicity. PKA inhibition by H89 reduced forskolin's protective effect against alcohol toxicity in a dose-dependent fashion. Thus, selective inhibition of PKA blocked the neuroprotective effect of forskolin.

Fig. 5. The nNOS gene is a downstream target of the cAMP pathway. Forskolin-induced expression of the nNOS promoter in human neuroblastoma cells. Each nNOS promoter beta-galactosidase plasmid was introduced into SKNSH cells, and the beta-galactosidase activity expressed from each construct was measured 48 h after transfection. Forskolin was added to corresponding cultures 12 h prior to harvesting. For all transfection assays, at least three independent experiments were performed. Beta-galactosidase activity is shown with standard deviation as error bars. Forskolin stimulated expression of the nNOS promoter–reporter construct, and this stimulation was substantially reduced when the two CRE sites were mutated.

Fig. 6. Forskolin increases the expression and activity of the endogenous nNOS gene, but not iNOS gene, in wild type CGN cultures. (A) CGN were treated with forskolin, and nNOS and iNOS gene expression levels were determined by real-time quantitative PCR assay in a time course experiment. nNOS and iNOS mRNA levels are presented, relative to the levels observed in untreated CGN. Values represent the mean of two independent data points. Error bars represent SEM. (B) Total nitrite levels in culture media were determined by the Griess reaction, as an indication of NOS activity. Values represent the mean of six independent data points. Error bars represent SEM. *Significantly different from control (p < 0.05).

Fig. 7. Adenoviral delivery of nNOS to CGN cultures leads to expression of nNOS at both the mRNA and protein levels. Cerebellar granule neurons derived from mice lacking nNOS (nNOS−/−) were infected with an adenoviral vector carrying the coding region of nNOS (Ad5-nNOS). Quantitative RT-PCR and Western blot analysis were carried out to determine the expression of nNOS messenger RNA (A) and protein (B). The amount of nNOS message is shown, relative to the amount in uninfected cells. An empty adenoviral vector carrying only a Bgl II site was used as a negative control for both messenger RNA and protein detection. β-Actin served as an internal control and indicates an equal amount of protein loading in each lane.

Fig. 8. Delivery of nNOS via an adenoviral vector protects CGN against alcohol-induced cell death. Cerebellar granule neurons derived from mice lacking nNOS (nNOS−/−) (A) and wild type (B) were infected either with an adenoviral vector carrying the coding region of nNOS (Ad5-nNOS) or with an empty vector (Ad5-Bgl II) 18 h after establishment of the cultures. Treatment groups were exposed to alcohol 24 h after the viral infection for an additional 24 h. (A) In uninfected cultures derived from mice lacking nNOS (nNOS−/−), alcohol induced a substantial amount of cell death. Delivery of nNOS via adenoviral vector blocked this toxic effect. In contrast, delivery of the empty vector (Ad5-Bgl II) had no protective effect. (B) In uninfected cultures derived from wild type mice, alcohol likewise induced cell death. Delivery of nNOS via adenoviral vector also protected these cells against alcohol toxicity, as it did in nNOS−/− cells. The experiments were repeated at least three times with different litters. Error bars represent standard error of the mean.